India cloudsummit Bangalore - Advanced Container Use-cases on AWS Container Services

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential
Advanced Use-Cases On AWS
Container Services
For India cloud summit, Bengaluru
Feb 15th 2019
Mani Chandrasekaran
Solutions Architect
Amazon Internet Services Pvt. Ltd.
http://twitter.com/cmani
http://www.linkedin.com/in/cmanikandan

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential
When the impact of change is small,
release velocity can increase
Monolith
Does everything
Microservices
Do one thing

Mobile
client
Client
IoT
Microservices architectures
API
Gateway
Account DB
Shipping DB
Inventory DB
Store Front
Web App
Account
Lookup
Shipment
Query
Inventory
Query

Cloud-native architectures are
small pieces, loosely joined

Containers are the best on ramp
towards modern applications

Why are enterprises
adopting containers?
• Accelerate software development
• Build modern applications
• Automate operations at web scale

Make AWS the BEST PLACE to run ANY
containerized applications

Helping customers scale containers
450+%
growth
Hundreds of millions
of containers started each week
of millions
of container instances

Why customers love AWS container services
Containers are a first-class citizen of the AWS Cloud
Deeply integrated
with AWS
Security and Compliance
Broad selection of compute instances
and IAM security, VPC networking,
load balancing, and autoscaling
ISO, HIPPA, PCI, SOC1, SOC2, SOC3
Infocomm Media Development Auth.
DevOps Workflow
Best place to build and operate
a complete DevOps workflow for
containers—AWS DevTools and Cloud9
DEV OPS

Typical use cases
• Microservices: Java, Node.js, Go, Web Apps, etc.
• Continuous Integration and Continuous Deployment (CICD)
• Batch Processing and ETL jobs
• Common PaaS Stack for Application Deployment
• Legacy Application Migration to the Cloud
• Hybrid Workloads
• AI/ML
• Scale Testing
• Backend for IoT use cases

AWS container services landscape
Management
Deployment, Scheduling,
Scaling & Management of
containerized applications
Hosting
Where the containers run
Amazon Elastic
Container Service
Amazon Elastic
Container Service
for Kubernetes
Amazon EC2 AWS Fargate
Image Registry
Container Image Repository
Amazon Elastic
Container Registry

Amazon EC2 Purchasing Options
On-Demand
Pay for compute capacity by the
second with no long-term
commitments
Spiky workloads, to define
needs
Reserved Instances
Make a 1- or 3-year commitment and
receive a significant discount off
On-Demand prices
Committed, steady-state
usage
Spot Instances
Spare EC2 capacity at savings of up
to 90% off On-Demand prices
Fault-tolerant, flexible,
stateless workloads

Why Spot Instances?
Inexpensive
Significant price savings of
up to 90% over On-Demand
Instances. Perfect for
stateless and fault-tolerant
workloads
Faster Results
Easily run multiple projects
simultaneously and speed up
job flows to generate
business results faster and
innovate faster
Easy Access
Launch spot instances via
the RunInstances function,
run-instances cmd, or from
AWS Console. Integrated
with other AWS and 3rd
party services
Resource Flexibility
Flexibility of ad-hoc
provisioning for multiple
instance types with an
option to hibernate, stop or
terminate instances when
reclaimed.

Provision Amazon EC2 Spot Instances
• Run Instances API
• Launch configuration / Auto scaling groups
API’s
• EC2 Fleet and/or Spot Fleet API
• RequestSpotInstances API

Amazon Elastic
Container Service

Scheduling and Orchestration
Cluster Manager Placement Engine
ECS

Containers + Spot = match made in heaven
ü Containers are stateless, fault-tolerant and a great fit for Spot Instances
ü Deploy containerized workloads and easily manage clusters at any scale at a fraction of
the cost with Spot Instances
ü Spot instances can be used with ECS or Kubernetes to run any containerized workload
Skyscanner is a travel fare aggregator
website and travel metasearch engine
based in Edinburgh, Scotland

Enhanced GPU support in Amazon
ECS
Blog - https://aws.amazon.com/blogs/compute/scheduling-gpus-for-deep-
learning-tasks-on-amazon-ecs/

Handson labs using ECS and Spot Instances
Amazon ECS Catsndogs workshop - https://github.com/aws-
samples/amazon-ecs-catsndogs-workshop

AWS Batch

Introducing AWS Batch
Managed
No software to install or
servers to manage.
AWS Batch provisions,
manages, and scales your
infrastructure
Integrated with AWS
Natively integrated with the AWS
platform, AWS Batch jobs can
easily and securely interact with
services such as Amazon S3,
DynamoDB, and Rekognition
Cost-optimized resource
provisioning
AWS Batch automatically
provisions compute resources
tailored to the needs of your jobs
using Amazon EC2 and Spot
Instances

Introducing AWS Batch
• Fully-managed batch primitives
• Focus on your applications (shell scripts,
Linux executables, Docker images) and
their resource requirements
• We take care of the rest!

AWS Batch Concepts
• Jobs
• Job Definitions
• Job Queue
• Compute Environments
• Scheduler

Job Definitions
Similar to ECS task definitions, AWS Batch job definitions specify how jobs are
to be run. While each job must reference a job definition, many parameters
can be overridden.
Some of the attributes specified in a job definition:
• IAM role associated with the job
• vCPU and memory requirements
• Mount points
• Container properties
• Environment variables
• Retry strategy
$ aws batch register-job-definition --job-definition-name gatk
--container-properties ...

Jobs
Jobs are the unit of work executed by AWS Batch as containerized
applications running on Amazon EC2.
Containerized jobs can reference a container image, command, and
parameters. Or, users can fetch a .zip containing their application and
run it on a Amazon Linux container.
$ aws batch submit-job --job-name variant-calling
--job-definition gatk --job-queue genomics

Easily run massively parallel jobs
Today, users can submit a large number of independent “simple jobs.”
AWS Batch also supports “array jobs” that run many copies of an
application against an array of elements.
Array jobs are an efficient way to run:
• Parametric sweeps
• Monte Carlo simulations
• Processing a large collection of objects

Demo !!!

Demo !!!
Run Monte Carlo Simulations on AWS Batch -
https://github.com/aws-samples/ec2-spot-montecarlo-
workshop

Demo !!!
Using EC2 Spot fleet

Demo !!!
Using a fully managed AWS batch

More use cases for AWS Batch ..
1. Fraud detection -
https://www.slideshare.net/AmazonWebServices/building-fraud-
detection-systems-with-aws-batch-and-containers-dvc301-aws-
reinvent-2018
2. Genome workflows -
https://aws.amazon.com/blogs/compute/building-high-
throughput-genomics-batch-workflows-on-aws-introduction-
part-1-of-4/
3. Data pipelines -
http://tech.adroll.com/blog/data/2018/08/08/running-jobs-
with-aws-batch.html
4. Much more …

Amazon Elastic
Container Service for Kubernetes

But where you run Kubernetes matters
Quality of the
cloud platform
Quality of the
applications
Your users

—CNCF survey

EKS is Kubernetes certified

Amazon EKS now available in
Mumbai !!

How are customer using Amazon EKS?
Microservices
PaaS
Platform-as-a-Service Enterprise App
Migration
Machine Learning

Amazon EKS architecture
Workers Workers Workers
Availability zone Availability zone Availability zone
etcd
Master Nodes
etcd
Master Nodes
etcd
Master Nodes

Auto scaling with Amazon EKS
• Two dimensions to scaling
• EC2 instance through K8s cluster AutoScaler
• Scale out Amazon EC2 Spot instances
• Pods through HPA
• Scale out pods

Auto scaling with EKS – Con’t …
Amazon EC2 Autoscaling
• Min
• Max
• Instance type
• ASG group
K8s Cluster Autoscaler
• Min, max, instance
type
• K8s node groups
• CA modes
• Auto-discovery
• Multi ASG
• Single ASG
• Master node
• HPA

EKS with Spot – sample architecture
VPC
Auto
Scaling
Auto
Scaling
Auto
Scaling
m4.large Spot
t2.medium Spot
On-Demand
Instances Instances Instances
Spot
Instance
Spot
Instance
Spot
Instance
Spot
Instance
Spot
Instance
Spot
Instance
Spot Interruption
Cluster Autoscaler

EKS with Spot Instance – Handson Labs
https://eksworkshop.com/spot

Machine Learning Frameworks on EKS
https://github.com/aws-samples/machine-learning-using-k8s

Kubeflow on Amazon EKS
https://aws.amazon.com/blogs/opensource/kubeflow-amazon-eks/

AWS Container Services Region availability
– as of Feb 15th 2019
Amazon ECS AWS Batch AWS Fargate Amazon EKS
Singapore, Tokyo,
Sydney, Seoul,
Mumbai, Beijing,
Ningxia, Ireland,
Frankfurt, London,
Paris, Stockholm,
N.Virginia, Ohio,
Oregon, N.California,
Montreal, Sao Paolo
N.Virginia, Ohio,
Montreal, Sao Paolo,
Ireland, Frankfurt,
London, Paris,
Singapore, Tokyo,
Sydney, Seoul,
Mumbai
Singapore, Tokyo,
Sydney, Seoul,
Mumbai, Ireland,
Frankfurt, London,
N.Virginia, Ohio,
Montreal
Singapore, Tokyo,
Sydney, Seoul,
Mumbai, Ireland,
Frankfurt, London,
Paris, Stockholm,
N.Virginia, Ohio,
Oregon
https://aws.amazon.com/about-aws/global-infrastructure/regional-product-
services/

THANK YOU
https://aws.amazon.com/containers
Mani Chandrasekaran
Solutions Architect
Amazon Internet Services Pvt. Ltd.
http://twitter.com/cmani
http://www.linkedin.com/in/cmanikandan

India cloudsummit Bangalore - Advanced Container Use-cases on AWS Container Services

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to India cloudsummit Bangalore - Advanced Container Use-cases on AWS Container Services

Similar to India cloudsummit Bangalore - Advanced Container Use-cases on AWS Container Services (20)

Recently uploaded

Recently uploaded (20)

India cloudsummit Bangalore - Advanced Container Use-cases on AWS Container Services